ES2998031T3 - Battery management system - Google Patents

Abstract

The invention relates to a battery management system (BMS) for accumulators, the batteries being able to consist of a single element or several single elements that can be arranged in a modular assembly and connected in series or in parallel, and by an electromagnetic or electronic disconnection device connected, on the one hand, to at least one pole of a cell and, on the other hand, to a terminal of the same polarity of a modular assembly or of a battery, the BMS comprising at least a single device for detecting deep discharges, overcurrent discharges and short-circuit discharges by measuring the voltage at the battery terminals. (Automatic translation with Google Translate, without legal value)

Description

DESCRIPTION

Battery management system

Technical field of the invention

The present invention relates to the field of lithium batteries for accumulators and, in particular, to the protection of said batteries.

State of the art

Storage batteries are composed of electrochemical elements that can be connected in series or parallel to obtain the necessary voltage and current.

Current batteries, for example, but not limited to those used in the aviation sector, which use lead or nickel-cadmium, have a high mass, short lifespan, and high self-discharge, and require regular maintenance. However, due to their low energy density, they pose little risk of overheating and fire. Often, a fuse or circuit breaker is sufficient to provide protection against short circuits.

In the event of a short circuit or overcurrent, it is necessary to disconnect the battery to avoid damaging it and preventing excessive heating of the battery or its connecting cables.

Lithium batteries also require protection circuits against overcurrents and short circuits. Typically, these circuits use electromechanical circuit breakers or fuses, or electronic circuits that measure current and control a switching device. However, measuring current is not straightforward.

There are several methods of current measurement (by shunt, magnetic measurement or by thermal effect).

In particular, document US-2019/302190 A1 is known, which describes a low-power electronic system for measuring both the charge measurement of a battery and the detection of overcurrents. The electronic system comprises a measuring circuit that measures the voltage of a battery, an interface circuit that transmits the battery voltage information to a processing device, and a control circuit. The measuring circuit comprises a counter and a voltage divider bridge placed in parallel with the battery, and whose division ratio is determined by a counter value. The measuring circuit also comprises a comparator that compares an output voltage of the resistance circuit with a reference voltage. When the power supply to the processing device is cut off, the counter sets the count value to a fixed value, and the control circuit protects the battery based on the comparator's comparison result.

Also known is document WO 2007/086971 A1, which describes a device for monitoring a rechargeable battery, the monitoring device comprising at least one current interruption switch for interrupting the current flowing through at least one associated cell, and a plurality of monitoring units for detecting the cell voltage. Each monitoring unit is associated with a single cell, and comprises a reference voltage unit for producing a defined reference threshold voltage, and a voltage comparison unit for comparing the reference threshold voltage with a partial cell voltage of the associated cell. The reference voltage unit is electrically powered from the voltage of the associated cell, and the voltage comparison unit is coupled to at least one current interruption switch for interrupting the current flowing through the associated cell.

However, these measurement methods can involve high power consumption, which may require the use of both a "standby" mode and an "active" mode. Furthermore, this is poorly compatible with short-circuit protection, which can occur at any time.

Additionally, the starter battery of a combustion engine must supply a very high current for a few seconds to a few tens of seconds. Therefore, the circuit breaker current (switching device) must be set to a fairly high value, on the order of half the short-circuit current (the maximum power supplied by a battery is reached when the voltage is half the open-circuit voltage, and the current is half the short-circuit current). With aging or at low temperatures, the internal resistance of the battery cells increases, so the short-circuit current decreases. This current may fall below the disconnection current. In this case, protection is no longer guaranteed. Use under these conditions can cause the battery to completely discharge its internal resistance, resulting in severe overheating and, subsequently, a fire.

Finally, for example, and without limitation, a non-modular 17 Ah battery can supply a short-circuit current of over 2000 A. Therefore, the disconnecting device that disconnects the battery from its operating circuit must be able to withstand this current. Semiconductors capable of withstanding this current are rare, and in practice, several lower-current components are connected in parallel. Current balancing is very difficult to achieve, requiring oversizing of components. Measuring a current of 2000 A also poses trade-offs between accuracy and static consumption.

Description of the invention

The present invention aims to overcome certain drawbacks of the prior art by proposing a method for detecting abnormal conditions and a lithium battery management system that is simple, reliable, and easily adaptable to different currents and voltages for individual or modular elements.

An objective of the invention is to enable the method to detect deep discharge, short-circuit discharge and overcurrent discharge conditions of lithium batteries, without going through a current measurement.

This objective is achieved by a method for detecting abnormal operating conditions (deep discharges, overcurrent discharges and short-circuit discharges) of an individual battery cell or a plurality of individual cells, comprising the following steps:

Sampling, at the common point of at least two resistors of a divider bridge with at least two resistors, of at least one voltage proportional to the voltage at the terminals of the individual element or of the set of individual elements;

compare the detected voltage with a reference threshold;

This comparison with the reference threshold triggers the implementation of:

an assessment of future developments by means of an integral, evaluated analogically or digitally;

a comparison of this evaluation with a detection volt threshold Td from which a disconnection of the single element or the set of individual elements monitored with respect to at least the battery terminals is carried out.

According to another feature, the digital integral evaluation step includes:

steps for calculating the slope of evolution (P) of the voltage curve using at least two measurements taken at least one step of comparing the calculated slope with a stored “RapidThreshold” value, i.e. if the slope exceeds the “RapidThreshold” value, applying a weighting coefficient that increases the acceleration of the evolution of the integral so that it crosses the activation voltage threshold Td more quickly, or if it is not exceeded, a weighting coefficient with no acceleration effect.

Another objective is to allow the management system to monitor deep discharge, short-circuit discharge, and overcurrent discharge conditions of lithium batteries, without having to go through current measurement.

This objective is achieved by a battery management system (BMS) for accumulators, said batteries being able to consist of a single element or of several individual elements that can be arranged in a modular assembly and connected in series, in parallel, or in a plurality of modular assemblies in series connected in parallel to form a battery; said system comprising suitable means for carrying out the steps of a detection method according to the invention.

According to another feature, the battery management system (BMS) comprises at least:

a voltage divider bridge with at least two resistors, to sample at least one voltage proportional to the voltage at the terminals of the single cell or the set of individual cells of the battery,

a detection device for detecting abnormal conditions,

a disconnect device,

said detection device communicates with the disconnection device to activate it in case of detection of abnormal conditions, said disconnection device being able to be connected to the battery and comprising at least two MOSFETs.

Another objective is to propose a digital solution for a detection device.

This objective is achieved by the battery management system (BMS) for accumulators according to the invention, which comprises a microprocessor equipped with at least one storage memory allowing to store at least one threshold variable of “Refintegration” and a stored detection voltage value T d; the memory also contains the program executed by the microprocessor, allowing to collect the points of the voltage curve, the comparisons and decisions, allowing the implementation of the integration equations, the microprocessor receiving as input the voltage V<global> that comes from the common point of the divider bridge between a resistance R1 and a resistance R2, and storing the measurements according to a determined frequency to observe the voltage curve V<global>, and compares the values of the voltage curve V<global> with the value of “Refintegration”, subsequently, when a crossing of the “Refintegration” threshold is detected, said threshold being defined by the value stored in the memory, activating the integration calculations of the V<global> curve, and comparing the values of the integration curve (V<integ.>) calculated with a stored detection voltage value Td to activate the disconnection device that performs the cut.

According to another feature, the microprocessor's storage memory also comprises the value of a variable “RapidThreshold” stored in order to determine, by comparing the variation dV of the voltage V<global> between two successive instants t1 and t2, with the “RapidThreshold”, whether the calculation of the integral of the voltage curve V<global> must take into account a weighting coefficient or not.

According to another feature, the calculation of the integral includes taking into account the variables of “Slope and/or Ordinance” calculated by the microprocessor from the data of the recorded voltage curve V<global>.

Another objective is to propose an analog solution for a detection device.

This objective is achieved by the battery management system (BMS) for accumulators according to the invention, which comprises at least approximately a comparator U1, a divider bridge (R1, R2, or R9, R4) mounted between the terminals of the modular assembly of the battery or of an individual cell of the battery whose common point with the resistors is connected to the input of the negative terminal of the comparator U1, to supply a voltage whose value is proportional to the voltage value V1 at the terminals of the battery, in the ratio defined by the values of the two resistors (R1, R2, or R9, R4), and the positive terminal of the comparator is connected to a diode or a supply cell to define the reference voltage V2.

Therefore, by changing the reference voltage and resistance values, the system can adapt to batteries with different voltage and current characteristics.

According to another characteristic, the integrating set comprises

a resistor R5 connected between the common point of the divider bridge R1, R2 and the negative input of comparator U1, and a set of resistor R8 and capacitor C1, mounted in series by means of a common terminal, connected by the other terminal of C1 to the output of comparator U1, the other terminal of R8 being connected to the common point of the two resistors R5, R8 and to the negative input of U1, adjusting the values of R5 and C1 to establish the disconnection intervention time before the battery deteriorates, in case of overcurrent detection.

According to another feature, a diode D2 is connected in parallel to the resistor R5, with its cathode connected to the common point of the divider bridge, to change the integration time constant of the integrator circuit in case of overcurrent or short circuit.

According to another characteristic, the comparator U1 has, at its output terminal, a voltage whose value characterizes a “non-conducting” state if the voltage applied to the input of the negative terminal of the amplifier U1 is greater than the value of the reference voltage V<2>, and a “conducting” state if the value of the voltage applied to the input of the negative terminal of the amplifier U1 is lower than the value of the reference voltage V<2>.

According to another feature, the detection device comprises a capacitor C3 mounted in parallel with R2 which, combined with R1, forms a filter to filter out high frequency disturbances.

According to another feature, in parallel with the R9 is mounted

a series assembly consisting of a resistor R3, a diode D3, with the cathode facing the positive terminal, and a Zener diode D4, with the cathode facing the common point of the divider bridge R9, R4,

a capacitor C5 that connects the common point of the bridge R9, R4 to the negative terminal of the battery or cell or modular assembly of individual elements.

According to another feature, a comparator circuit U2 with hysteresis, arranged downstream of the comparator circuit U1, comprises a hysteresis assembly around the amplifier U2, which receives, at the input of its negative terminal, the voltage value of the output of the amplifier U1.

According to another characteristic, the hysteresis comparator U2 comprises resistors R3, R4 mounted as a divider bridge between the positive and negative terminals of the battery, and whose common point to R3 and R4 is connected to the positive input of the comparator U2, and a resistor R6 of which connects the output of U2 to its positive input to define the threshold and the hysteresis of the hysteresis comparator circuit comprising the amplifier U2.

According to another feature, the detection device comprises a resistor R7 connected to the positive terminal of the battery, in series with a diode D1, in the forward direction during normal operation, and a capacitor C2 connected on one side to the cathode of the diode and, on the other side, to the negative terminal of the battery to allow it to be charged during normal operation; the supply inputs of the two comparators U1, u2 are connected to the common point of D1 and C2, this common point being used to maintain the supply of the amplifiers U1 and/or U2 when the battery voltage is reduced after a short circuit, to allow the disconnection to be activated.

According to another feature, the reference voltage V2 at the positive input of the comparator U1 is provided by a Zener diode D2, said Zener diode D2 being connected by means of a resistor R1 to the common point of D1 and C2, the cathode of the Zener diode D2 also being connected by means of a capacitor C1 to the negative terminal of the battery or modular set of elements.

According to another feature, the hysteresis comparator U2 comprises a capacitor connected in parallel with a resistor R4 and which, combined with another resistor R3 or R2, forms a filter to filter out high frequency disturbances and establish a minimum disconnection time.

According to another feature, the positive input of the hysteresis comparator U2 is connected via a resistor R2 to the common point of R3, R9 and R7.

According to another feature, the detection device comprises a flip-flop connected to the output of U1 or U2 to store each action of the detection device after each detection of deep discharges, overcurrent discharges and short-circuit discharges.

Another objective is to propose a disconnection device, usable with the detection device, formed by electronic components associated in an assembly that protects the components of the disconnection device. This other objective is achieved by the disconnection device, which comprises a switching device (30) in which

a first MOSFET M1 is connected by its source to the negative terminal of a set of individual elements, said MOSFET M1 receiving, at its gate, the voltage source that drives M1, said source supplying a voltage chosen so that M1 is on,

a Zener diode D3, connected in opposition between the gate and source of M1, and a capacitor C2 protect the gate of the MOSFET from excessively high or high frequency voltages, and

a Zener diode D1 mounted opposite the gate of M1 and the drain and with a resistor R3 and a diode D2 in the forward direction in the drain-grid direction limit the switching speed of M1, and

a circuit consisting of a Schottky diode D4 mounted opposite to the drain of M1 and in series with a capacitor C1 and a resistor R1 connected to the positive terminal of the battery (4) to limit the overvoltage when opening M1, in parallel with the Schottky diode D4 there is mounted a fixed resistor I1 connected, on the one hand, to the cathode of the diode and, on the other hand, to the drain of a second MOSFET M2 whose source is connected to the anode of the Schottky diode D4, the gate of M2 being controlled by an output of the detection circuit, so as to prevent charging.

According to another characteristic, the disconnection device comprises a second switching device in which a first MOSFET M1 connected by its source to the negative terminal of a set of individual elements, the gate of this MOSFET M1 is controlled by a voltage, this source supplying a voltage chosen so that M1 is always on,

a circuit consisting of a Schottky diode D4 mounted opposite to the drain of M1, in parallel to the diode D4 a fixed resistor I1 is mounted connected, on the one hand, to the cathode of the Schottky diode D4 and, on the other hand, to the drain of a second MOSFET M2 whose source is connected to the anode of the Schottky diode D4, the gate of M2 being connected to the positive terminal of said set of individual elements,

a Zener diode D6 and a resistor R6 in series with the gate of M2, with the Zener diode D6 mounted in the forward direction in the drain-gate direction,

a Zener diode D5 mounted opposite the gate and source of M2 to define, with the Zener diode D6, the value of the voltage at the gate of M2, and at which M2 is on,

a capacitor C5 connected between the gate and source of M2, and in parallel with Zener diode D5, to protect the gate of M2 from high frequency voltages,

an optocoupler OP1 mounted between the gate and source of M2, and in parallel with capacitor C5, to block M2 in case the voltage or temperature of one element of the array of individual elements is exceeded, then the MOSFET M2 cutting off the load current.

Other features and advantages of the present invention are detailed in the following description.

Brief description of the figures

Other features and advantages of the present invention will be more clearly understood upon reading the following description, made with reference to the accompanying drawings, in which:

[Figure 1]

Figure 1 shows a circuit diagram of the discharge detection device (2) for detecting deep discharges, overcurrent discharges and short-circuit discharges, according to an embodiment;

[Figure 2]

Figure 2 shows a circuit diagram of said detection device according to a variant;

[Figure 3]

Figure 3 shows a diagram of an example of a circuit for measuring current through a known resistor from the prior art, and which has the known drawbacks;

[Figure 4A]

Figure 4A shows the visualization of the changes in voltage at the terminals of comparators U1 and U2 in case of overcurrent, according to an embodiment for a 24.4 volt battery and a detection or activation voltage Td of 16 volts;

[Figure 4B]

Figure 4B shows the response of a digital integrator array operating according to the flow diagram of Figure 4D, according to an embodiment used with a 16 volt battery and a detection or activation voltage Td of 12 volts;

[Figure 4C]

Figure 4C shows the response of an analog integrator assembly, according to another embodiment, used with a 16 volt battery and a detection or activation voltage Td of 12 volts;

[Figure 4D]

Figure 4D shows a flowchart explaining the program for calculating the response of a digital integrator array, according to one embodiment, with parallelism of analog embodiments;

[Figure 4E]

Figure 4E shows the short-circuit response of an analog integrator assembly, according to one embodiment, used with a battery of approximately 14 volts and detection or activation voltage Td of 14 volts, in the event of a short circuit;

[Figure 4F]

Figure 4F shows the short-circuit response of a digital integrator assembly operating according to the flow diagram of Figure 4D, according to one embodiment, used with a battery of approximately 14 volts and a detection or activation voltage Td of 10 volts;

[Figure 4G]

Figure 4G shows the slow discharge response of an analog integrator assembly, according to one embodiment, used with a battery of approximately 14 volts and a detection or activation voltage Td of 14 volts;

[Figure 4H]

Figure 4H shows the slow discharge response of a digital integrator assembly operating according to the flow diagram of Figure 4D, according to one embodiment, used with a battery of approximately 14 volts and a detection or activation voltage Td of 10 volts;

[Figure 4I]

[Figure 4I] shows the response, in the case of discharge at V/2, of an analog integrator assembly according to an embodiment used with a battery of approximately 14 volts and a detection or activation voltage Td of 14 volts to determine the maximum current of the battery;

[Figure 4J]

Figure 4J shows the response, under discharge at V/2, of a digital integrator assembly operating according to the flow diagram of Figure 4D, according to one embodiment, used with a battery of approximately 14 volts and a detection or activation voltage Td of 10 volts to determine the maximum battery current;

[Figure 5]

Figure 5 shows a circuit diagram of the first switching device upon discharging of the disconnecting device, according to one embodiment;

[Figure 6]

Figure 6 shows a circuit diagram of the second switching device upon loading of the disconnecting device, according to one embodiment;

[Figure 7]

Figure 7 shows a diagram of the interconnection between the battery management system (BMS) and the disconnect device, according to one embodiment;

[Figure 8]

Figure 8 shows the diagram of a battery comprising a BMS and a disconnection device, according to one embodiment,

[Figure 9]

Figure 9 shows in the diagram a disconnection characteristic, obtained by the detection device, as described and very close to that of a magnetothermal circuit breaker of the state of the art, according to an embodiment.

Description of the achievements

The present invention relates to a method and a battery management system (1) (BMS) for accumulators. Said batteries may consist of a single element or of several single elements or cells, which may be arranged in a modular assembly connected in series, in parallel or in a plurality of modular assemblies in series associated in parallel to constitute a battery (4), and an electronic disconnection device (3) connected, on the one hand, to at least one pole of a cell and, on the other hand, to a terminal of the same polarity of a modular assembly or of a battery (4).

According to another variant, the battery management system (1) (BMS) for accumulators comprises at least one device (2) for detecting deep (or slow) discharge, overcurrent discharge and short-circuit discharge, in each individual element or modular assembly of the battery (4), which comprises at least one BMS device (Figure 8 illustrates a battery with a BMS device), said system (BMS) being characterized in that the detection device (2) is unique.

The detection device (2) comprises an analog or digital comparator (Un) U1 that directly compares a voltage that is proportional, in a given ratio, to the voltage of the individual element or the modular assembly, without using a resistive shunt, with a reference voltage (V2 or Refintegration) and evaluates the voltage variations to activate or not activate the disconnection of each individual element or modular assembly of the battery (4), according to a detection voltage Td from which the disconnection device is actuated.

The proportional relationship between the measured voltage and the reference voltage corresponds to the relationship between the reference voltage (V<2> or Refintegration) and the detection or disconnection voltage Td from which the disconnection device is actuated.

Figure 3 illustrates an example of current measurement using a shunt resistor. This measurement involves placing a shunt in series across the line. The total voltage V<tot.> is given by V<tot.>= V<battery>- V<rShunt>, with V<rShunt>= R * I, the value across the shunt resistor terminals. Therefore, the higher I, the higher the voltage drop. In this case, the consumption of a comparator U1 is very high, 1 mA for a differential amplifier versus less than 10 pA for voltage measurement.

Furthermore, the voltage V<battery> measured across the battery terminals or each cell is of the form V<battery> = ER * I, where E is the open-circuit voltage, R is the internal resistance, and I is the current supplied by the battery. E and R are known for a given battery, and the voltage V<battery> provides access to I, the supplied current. The maximum power that a battery is capable of supplying is close to (E / 2) * (I<cc>/ 2) (where E is the open-circuit voltage and I<cc> is the short-circuit current). Since the internal resistance tends to increase with battery aging, the maximum power decreases with battery aging. Disconnection by monitoring the voltage V<battery> instead of the current offers optimal protection for the battery throughout its lifetime. It is then evident that detection by voltage measurement is advantageous.

Furthermore, in the event of a failure of one or two lines comprising a set of individual elements or cells of the battery, the detection device of the battery management system (1) (BMS), as described according to the invention, is configured (in particular, by detecting by voltage measurement) to self-adapt the threshold or activation voltage Td so as to detect abnormal conditions (deep discharge, short-circuit discharge, overcurrent discharge) of the battery. In other words, the detection device of the battery management system (1) makes it possible to protect the battery, regardless of its aging state.

The individual accumulator elements of a battery (4) that must be protected may have, for example, and in a non-limiting manner, the following characteristics:

Open circuit voltage = 3.3 V

Internal resistance = 0.012 ohms (with connections)

Maximum current = 70 A; 120 A 10 s

End-of-discharge voltage = 2 V

End-of-charge voltage = 3.6 V

Maximum voltage = 4 V

A battery (4) can be modeled by a perfect voltage source in series with an internal resistance. In the exemplary case of an “8S1P” array of 8 individual cells in series, for example, and not limited to, there is an open charge (or end-of-charge) voltage of 26.4 V and an internal resistance of 0.1 ohm (with connections). The short-circuit current can therefore reach 264 A, while the specified maximum current is 120 A for 10 s. Overcurrent protection is therefore required.

The claimed fault (or abnormal condition) detection circuit diagram, illustrated in Figure 1, comprises a voltage divider R1 and R2, a voltage reference V2 of 1.24 V, for example and not limited to, and V1 being the voltage across the battery terminals monitored by the device. In the above example, the voltage at the end of discharge must not fall below 2 V per cell, i.e. 16 V for 8 cells in series. It is then possible to choose V1 = 16 V as the value of the detection or activation voltage Td. For this value of the detection or activation voltage, the division ratio of the voltage dividers R1 and R2 will be chosen equal to V2 / Td; i.e. 1.24 V / 16 V = 0.0775. Therefore, in static operation, when the battery voltage (4) is higher than the detection or activation voltage Td = 16 V, the voltage applied to the input - of the amplifier U1 is higher than 1.24 V, its output is in the “ non-conducting” state, the output of the comparator U2 is in the “ conducting” state, and the operation is “ normal” (no disconnection).

When the battery (4) is slowly discharged (deep discharge) and its voltage drops below 16 V, the voltage applied to the input - of amplifier U1 is less than 1.24 V, its output goes into the “ conducting” state, the output of comparator U2 changes to “ non-conducting”, and the disconnection is activated. The output of comparator (or amplifier) U1 is the integral of the voltage to be monitored according to an analog implementation (Figure 4C) or another digital implementation (Figure 4B). Therefore, as shown in these figures, if the voltage V<global> changes linearly, the output of the integrator will be a second degree function. If the voltage variation is a line of the form ax b, then the output is a function of the type ax<2>+ bx c.

In the expression for the voltage across the terminals of a battery or a single battery cell given by V<battery>= ER * I, it is E, the open circuit voltage, which decreases when the battery is fully discharged.

When the battery (4) delivers a high current (overcurrent), for example, 120 A, its voltage drops rapidly to 14.4 V (curve V1, Figure 4A). The above assembly detects this situation.

In the event of a short circuit, voltage V1 drops very quickly to a very low value, and the voltage across the terminals of R2 approaches 0 V; this situation is also detected.

This same assembly could be used in another variant of the embodiment of Figure 2 (assembly with Zener diode D4) by replacing resistors R1, R2 with resistors R9, R4.

The detection device, as described in the present application, allows the detection of voltage variations with a non-linear progressiveness, such that a strong overcurrent causes a disconnection of the battery after a period of between 10 ms and 100 ms, and such that a low overcurrent causes a disconnection in a time of between 10 s and 100 s.

An overcurrent or short-circuit causes the battery components to overheat, which can cause a fire. Heat dissipation is proportional to the square of the current, so the disconnection must be faster the more significant the excess current. A fuse or a magnetothermal circuit breaker cuts off the current in a time that is all the shorter the more significant the excess current. This involves setting a maximum current and low-battery protection in case of aging of the battery components, which cannot withstand a certain maximum current value due to the increase in the battery's internal resistance with aging. In other words, in case of aging, the fixed maximum current cannot be reached, and battery detection and protection will be ineffective. In contrast, in the present invention, which relates in particular to an overcurrent detection device based on a measurement of the voltage evolution across the battery terminals as a function of time, there is no restriction related to setting a maximum current value. If the measured voltage is below a certain value, corresponding to a low current, the disconnection time will be long, as illustrated in Figure 9. If the measured voltage is below a certain value, corresponding to a high current or a short circuit, the disconnection time will be short (see Figure 9). This protects the battery regardless of the aging of its components.

The detection functionality using battery voltage measurement combined with the use of an electronic circuit breaker makes it possible, in particular, to emulate the behavior of a thermal-magnetic circuit breaker without the restriction of setting a maximum current to improve battery safety. For example, Figure 9, which illustrates a disconnection curve obtained using an electronic circuit breaker according to the invention, is similar to that of a disconnection curve obtained using a shunt measurement and a thermal-magnetic circuit breaker, albeit with much greater accuracy.

The embodiments provide additional protection for circuit operation, to preserve component life or to avoid having to select expensive components because they meet high specifications. These embodiments may not be fully compatible with each other or with the main embodiment.

Therefore, the detection device (2) comprises a voltage divider formed by at least two resistors R1 and R2, the voltage divider being connected to the terminals of the battery (4) to reduce the voltage at the input of the comparator U1.

Similarly, a resistor R5 is connected between the common point of the divider bridge R1, R2 and the negative input of comparator U1, and a resistor R8, the capacitor assembly C1 mounted in series by a common terminal, is connected by the other terminal of C1 to the output of comparator U1, and the other terminal of R8 is connected to the common point of the two resistors R5, R8 and to the negative input of U1.

In another embodiment, diode D2 is connected in parallel to resistor R5, with its cathode connected to the common point of the divider bridge. In this configuration, diode D2 allows the integration time constant of the integrator assembly to be changed in the event of a very large or high current input, for example, in the case of an overcurrent or a short circuit. This change in the integration time constant is equivalent to the use of weighting in the signal (voltage) integration in the case of a digital integrator (see below).

Thus, in case of overcurrent, illustrated in Figure 4A, the battery (4) supplies a high current, for example 120 A, and its voltage drops rapidly to 14.4 V (V<1>, Figure 4A). The voltage difference between 26.4 V and 14.4 V is integrated by the integrator circuit R5-R8-C1 (in the integral mathematical sense) connected to the terminals of the amplifier U1, with an integration constant that depends mainly on R5-C1. Its output (V<integr.>, [Figure 4A]) slowly changes from “0 V” to “short-circuit voltage Vcc” and the hysteresis comparator U2 switches (V<3>, [Figure 4A]) when its threshold (V<4>, [Figure 4A]) of detection or activation voltage Td is reached. The disconnection is activated. The time constant R5-C1 is set so that the disconnection occurs before 10 s to 12 s, in this case.

During a short circuit, voltage V1 drops very quickly to a very low value, and the voltage across R2 approaches 0 V. Diode D2 conducts, the voltage at the - input of U1 is 0.6 V (the input is still at 1.24 V), and the output of U1 quickly changes to “Vcc”. The output of comparator U2 goes to “0” and the shutdown is activated.

In all these cases, a flip-flop stores this action and it is necessary to “reset” (i.e. reinitialize) the battery (4) by pressing a “start” button accessible from the outside of the battery casing.

Capacitors C3 and/or C4 filter out any high-frequency disturbances and establish the minimum disconnection time.

The principle of measurement via an integrating circuit, as described in this application, is a principle of measuring a global voltage that allows a current value to be tracked. This principle is only valid in the field of batteries when the internal resistance of the voltage generator is known. In this case, and only in this case, said integrating assembly can be used in analog (as shown in Figure 1) or digital form.

For example, and not limited to, the response or output of a digital integrator can be calculated as follows:

A voltage variation is considered represented by x = (-0.25 * V<global>+ 2.5) * weighting, where V<global> is a voltage obtained from the battery voltage by using a voltage divider bridge (R1-R2, or R9-R4), and weighting is a variable that allows changing the integration constant. The above equation can be modified depending on the accumulators used.

The output or response of the digital integrator with the general form y = Integration(x), where Integration( ) represents the integral calculus, can be calculated using either as a first progressive equation which consists of taking the value of x, defined above, and raising it to an even power (2, 4, 6, 8, etc.), for example, y = x<2>.

To get even closer to the analog integrator (Figure 4A, Figure 4C), a second progression equation can be used, defined, for example, and not limited to, Y = Rate * (-ln(x)), where Rate is an integration constant expressed in seconds. This equation makes it possible to imitate the behavior of a capacitor whose voltage at its terminals evolves exponentially.

Figure 4D shows a diagram for calculating the response of a digital integrator according to the second progressiveness equation. Each calculation step represents the components of the sensing device that may be involved in the calculation operations. The diagram can be divided into three phases: a measurement (PM) and comparison phase, an integration phase (PI), and a dropout phase (PD).

In the measurement phase (PM), the voltage divider bridge R1-R2 (or R9-R4) allows a measurement V = V<global> of the voltage at the input of the sensing device to be determined from the battery voltage V1.

The variable “Refintegration” is the integration reference and corresponds to a voltage value below which the input signal V will be integrated. If the voltage V is greater than the variable “Refintegration”, the battery is in a normal operating situation. If V is lower than the variable “Refintegration”, the battery is functioning abnormally, and the process that can lead to the disconnection of said battery is activated. This variable “Refintegration” is therefore equivalent to the reference voltage V<2>. The integration phase then begins, where the integrator response must be calculated.

If the voltage V is lower than the variable “ Refintegration” , the program activates either the use of a normal integration constant in the calculation performed, or the use of a weighting for the integration constant. This weighting, as represented in the PI table, is used if the voltage is lower than a second comparison variable, called “ RapidThreshold” , which allows defining a voltage threshold from which the “ weighting” variable (defined above) is used or not in the calculation of the voltage variation. For example, and not limited to, the voltage variation has a general form of type x = (slope * V<global>+ ordinate) * weighting.

If the difference or variation of the input voltage V, dV, between a time t1 and a time t2 (or between two successive measurements of the voltage V), defined by dV = |V(t2) - V(t1)|, is greater than the variable “ RapidThreshold” , the variable “ weighting” takes the value 5, for example. If, on the contrary, said difference or variation of the input voltage V, dV, is less than the variable “ RapidThreshold” , the variable “ weighting” takes the value 1. Which corresponds to the use of a normal integration constant.

The voltage measurement time interval may be, for example, and not limited to, between 1 ms and 100 ms. The value of the variable "RapidThreshold" may be defined according to the measurement time interval and by monitoring the voltage variation between two times t1 and t2, corresponding to said time interval, used to perform the voltage measurements, in order to improve the conditions for detecting abnormal conditions. For example, and not limited to, for Figure 4B, the measurement time interval used is 10 ms, and the "RapidThreshold" value is 0.01 volts. This corresponds to a voltage drop dV = 0.01 volts every 10 ms.

The variables “Ordinate” and “Slope”, which are obtained by storing the measurement points and calculating them, for example, by fitting the stored voltage data or by using two points of the stored voltage curve between two times t1 and t2 to deduce the “slope” (for a linear voltage variation) and then the “ordinate”, allow to define the voltage variation. In the example where x = (-0.25 * V<global>+ 2.5) * weighting, the slope is -0.25 and the ordinate is 2.5.

The step of comparing the voltage variation dV is equivalent to a step of comparing the calculated slope with the stored “RapidThreshold” value, i.e. if the slope exceeds the “RapidThreshold” value, applying a weighting coefficient (e.g. 5) that increases the acceleration of the evolution of the integral so that it exceeds the activation voltage threshold Td more quickly, or if it is not exceeded, a weighting coefficient with no acceleration effect (e.g. 1).

Once the voltage variation is obtained, the signal is integrated according to the second progressive equation, for example. The output signal therefore corresponds to the integration of the input signal.

The variable “Progressiveness coefficient” corresponds to an integration constant (Rate in the second progressiveness equation).

In the embodiment using a digital integrator, those skilled in the art will understand that the assembly using comparators U1 and U2 is replaced by a microprocessor acting as a digital comparator (Un). Said microprocessor is equipped with a storage memory that allows storing the threshold variables “Refintegration” and “RapidThreshold”, and the calculation variables “Ordinate” and “Slope” defined according to these thresholds. The memory also contains the calculation program that allows collecting the points of the voltage curve (V<global>, etc.), the comparisons and decisions, the implementation of the equations, the integration and the decisions represented in the flowchart of Figure 4D. As input, the digital circuit only receives the voltage V<global> from the common point of a divider bridge between a resistor R1 and a resistor R2, and makes measurements according to a determined frequency to observe the voltage curve V<global>, then from the detection of the crossing of the “Refintegration” threshold which, in the example of Figure 4B, is chosen to be less than 3 volts per cell element, or 12 volts for a battery of 4 cell elements in series, from this reference voltage V<2>, the microprocessor program activates the calculations to obtain the comparison with the variable “RapidThreshold” of the variation dV of the voltage V<global> between two successive instants t1 and t2 (or between two successive measurements), in order to determine whether or not to use a “Weighting” variable. Therefore, in the case of a start-up that causes a significant voltage drop from 14 to almost 6 volts, the variable “RapidThreshold” is set, for example, and not limited to, 0.01 volts in the example of Figure 4B. The variable “RapidThreshold” will be crossed and the integration will be performed with weighting to avoid an excessively rapid cut-off that would prevent the start. In the diagram shown in Figure 4<b>, it is observed that the battery voltage, which has dropped rapidly to almost 6 volts, and remains constant for approximately 18 seconds, the digital circuit integrates the constant value into a straight line, which remains below the detection or activation voltage Td, which is chosen at 1 volt. The integrator response or output voltage can be obtained, for example and without limitation, with a program such as the one defined in the appendix of this application, where the variable “GeneralVoltage” corresponds to the voltage V<global> at an instant t1 = t, and the variable “LastGeneralVoltage” represents the value of the voltage V<global> at the instant t2 = t - 1. The variable “ORDINATE_ORIGIN” corresponds to the variable “Ordered” defined above, and the variable “LastIntegratedValue” corresponds to the integral calculation or the integrator response.

The calculation of the integral, or the integrator response, may involve taking into account the “Slope and/or Ordinate” variables calculated by the microprocessor from the data of the recorded voltage curve V<global>.

Integration is activated as soon as the overall voltage V<global> falls below V<2>= Refintegration = 9 volts.

Subsequently, during use, the battery voltage suddenly drops from 14 volts to approximately 9 volts, and then slowly decreases over time along a straight line to 6 volts. The ordinate of the line is approximately 2.3 volts and the slope is lower than before, and the variation dV in voltage between two successive measurements can be greater (depending on the value of the slope) than the variable “RapidThreshold” (for example, 0.01 volts in the example shown in Figure 4B).

When the value at the integration output reaches the threshold corresponding to the detection or activation voltage Td of 1 volt, the cut-off is activated.

Finally, in the digital version or variant, during a short circuit, the Vgiobai voltage drops very quickly to a very low value, a short circuit detection threshold is stored and, as soon as the processor detects that this threshold is crossed, it activates the disconnection signal.

In Figure 4B, which illustrates the response or output signal of a digital integrator according to the example described above, the digital integrator shows behavior similar to that of an analog integrator (Figure 4C) in the time interval between t = 40 s and approximately t = 120 s.

In the turn-off phase, the response calculation is used to check whether a turn-off should be triggered or not. The turn-off is triggered when the integrator response is greater than a given threshold corresponding to the detection or trigger voltage Td. In the example above, illustrated by Figure 4B, Figure 4C, and Figure 4D, this threshold is set to approximately 1 or 1.24 volts. For example, and not limited to, the threshold value can be normalized to 1.

Figure 4E and Figure 4F illustrate, respectively, the response of an analog integrator and a digital integrator when a short circuit is detected. The measurement time interval is set to 10 ms, and the “RapidThreshold” variable is set to 0.1 V.

In the example of Figure 4F, which shows the response of a digital integrator, a drop in the voltage Vglobal at the battery terminals can be observed from a value of approximately 14 V to a value close to 0 V, in a time interval between 0.2 s and 0.3 s, and where the digital integrator circuit integrates the input signal with a weighting value = 150, in order to increase the acceleration of the evolution of the integral. This weighting is performed so that the output signal or the response crosses the trigger threshold Td = 10 V as quickly as possible (which is set to 14 V, in the case of the analog integrator). This Td value can be normalized to 1, as mentioned above. From 0.3 s to about 4.2 s, the value of Vglobal remains constant and close to 0 V, and the response of the digital integrator circuit is a straight line that passes above the turn-off trigger voltage Td after a period of about 1 s. When the battery returns to normal operation, after 4.2 s, the response of the digital integrator circuit is zero. The behavior of the digital integrator is similar to that of the analog integrator circuit illustrated in Figure 4E.

Figure 4G and Figure 4H illustrate, respectively, the response of an analog integrator and a digital integrator when a slow or deep discharge is detected. The measurement time interval is also set to 10 ms, and the value of the “RapidThreshold” variable is equal to 0.1 V.

As illustrated in Figure 4H, during slow or deep discharge, the input signal or voltage Vglobal changes to a straight line descending from 14 V to approximately 5 V, in a time interval between 0 and 116 s. The digital integrator circuit (or digital integrator) integrates the input signal, taking into account a weighting value equal to 150. The disconnection process is activated when the response curve is greater than the value of the activation voltage Td, set in this example at approximately 10 V (which is set to approximately 14 V in the case of the analog integrator). As in the cases of detection of abnormal conditions such as an overcurrent (Figure 4B) or a short circuit (Figure 4F), the behavior of the digital integrator is also similar to that of the analog integrator circuit illustrated in Figure 4G, in case of slow discharge.

Therefore, those skilled in the art will understand that the battery management system, as described in the present application, is suitable for detecting abnormal conditions comprising at least overcurrent, short circuit, and slow or deep discharge. This detection is carried out without having to change an electronic component of the detection device, which is unique, or having to integrate a new electronic component based on the abnormal conditions to be detected. Another advantage is that the detection device does not need to comprise as many electronic circuits as there are abnormal conditions to be detected. This makes it possible to avoid overloading the management system, which can cause maintenance problems in the event of a breakdown (in particular, in tracing the source of the breakdown).

Figure 4I and Figure 4J illustrate, respectively, the response of an analog integrator and a digital integrator to a V/2 discharge. The V/2 discharge, which occurs over a short period, for example, about 15 s, makes it possible to determine the maximum current admitted by the battery.

In certain embodiments, the battery management system (1) for accumulators comprises at least approximately a comparator U1, a divider bridge (R1, R2, or R9, R4) mounted between the terminals of the modular battery assembly (4) or of a single battery cell (4) whose common point with the resistors is connected to the input of the negative terminal of the comparator U1, to supply a voltage whose value is proportional to the voltage V1, in the ratio defined by the values of the two resistors (R1, R2 or R9, R4), and the positive terminal of the comparator is connected to a diode or a supply cell (not shown) to define the reference voltage V2.

In certain embodiments, the integrating assembly comprises a resistor R5 connected between the common point of the divider bridge R1, R2 and the negative input of the comparator U1, and a set of resistor R8 and capacitor C5, mounted in series by means of a common terminal, is connected by the other terminal of C5 to the output of the comparator U1, and the other terminal of R8 is connected to the common point of the two resistors R5, R8 and to the negative input of U1, the values of R5 and C1 are adjusted to establish the disconnection intervention time before the deterioration of the battery (4) in case of overcurrent detection.

In certain embodiments, comparator U1 has, at its output terminal, a voltage whose value characterizes a “non-conducting” state if the voltage applied to the input of the negative terminal of amplifier U1 is greater than the value of the reference voltage V2, and a “conducting” state if the value of the voltage applied to the input of the negative terminal of amplifier U1 is less than the value of the reference voltage V2.

In certain embodiments, the detection device (2) comprises a capacitor C3 mounted in parallel with R2 which, combined with R1, forms a filter for filtering out high frequency disturbances.

In certain embodiments, mounted in parallel with R9 is a series assembly consisting of a resistor R3, a diode D3 with the cathode facing the positive terminal, and a Zener diode D4 with the cathode facing the common point of the divider bridge R9, R4, a capacitor C5 connecting the common point of the bridge R9, R4 to the negative terminal of the battery (4) or of the modular array of individual elements. The pair (R3, C5) may be configured to obtain a fast time constant, and the pair (R9, C5) may be configured to obtain a slow time constant.

In certain embodiments, a comparator circuit U2 with hysteresis, arranged downstream of the comparator circuit U1, comprises a hysteresis array around the amplifier U2, which receives, at the input of its negative terminal, the voltage value of the output of the amplifier U1.

In certain embodiments, the hysteresis comparator U2 comprises resistors R3, R4 mounted as a divider bridge between the positive and negative terminals of the battery (4) V1, and whose common point to R3 and R4 is connected to the positive input of the comparator U2 and a resistor R6 of which connects the output of U2 to its positive input, so as to define the threshold and the hysteresis of the hysteresis comparator circuit comprising the amplifier U2.

In certain embodiments, the detection device (2) comprises a resistor R7 connected to the positive terminal of the battery (4), in series with a diode D1, in the forward direction in normal operation, and a capacitor C2 connected, on the one hand, to the cathode of the diode and, on the other hand, to the negative terminal of the battery (4), to allow it to be charged during normal operation; the supply inputs of the two comparators U1, U2 are connected to the common point to D1 and C2, and this common point is used to maintain the supply of the amplifiers U1 and/or U2 when the voltage of the battery (4) collapses following a short circuit, to allow the disconnection to be triggered.

In certain embodiments, the reference voltage V2 at the positive input of comparator U1 is supplied by a diode D2, which is a 16 V Zener. This diode is also connected by a resistor R1 to the common point of D1 and C2. The cathode of diode D2 is also connected by a capacitor C1 to the negative terminal of the battery (4) or to the modular array of elements. Diode D2 can also be replaced by a voltage reference.

In certain embodiments, the hysteresis comparator U2 comprises a capacitor (C4, Figure 1, or C3, Figure 2), connected in parallel with a resistor (R4, Figure 1, or R3, Figure 2) and which, combined with another resistor (R3, Figure 1, or R2, Figure 2), forms a filter for filtering out high frequency disturbances and setting a minimum disconnection time.

In certain embodiments, the positive input of hysteresis comparator U2, Figure 2, is connected via a resistor R2 to the common point of R3, R9, and R7.

In certain embodiments, the detection device (2) comprises a flip-flop connected to the output of U1 or U2 to store each action of the detection device (2) after each detection of deep discharges, overcurrent discharges and short circuit discharges.

In certain embodiments, the battery management system (BMS) (1) communicates via the detection device (2) with a disconnection device (3) contained in the battery. The disconnection device is connected, on the one hand, to the negative pole of each modular block or of each battery and, on the other hand, to the negative terminal of the battery (4), and uses at least two MOSFETs (M1, M2).

In another embodiment, the battery management system (1) (BMS) comprises, as illustrated in Figure 7, a disconnection device (3) that communicates with the detection device (2). The disconnection device is connected, on the one hand, to the negative pole of each modular block or of each battery and, on the other hand, to the negative terminal of the battery (4), and uses at least two MOSFETs (M1, M2).

The disconnection device (3) comprises a switching device (30, 31) which may be electromechanical, a relay, for example, or a semiconductor.

Ideally, the disconnect device should have very low static power consumption both in the on (conducting) state and in the off (non-conducting) state. For example, and not limited to, a bistable relay or a MOSFET transistor meet this requirement.

In certain embodiments, the disconnect device (3) comprises a first switching device (30) in which

a first field-effect transistor (MOSFET) M1 connected by its source to the negative terminal of a set of individual elements, this MOSFET M1 receiving, at its gate, the voltage source controlling M1 (which comes either from the output of the comparator U1, in the variant of Figure 2, or from the output of the hysteresis comparator U2, in the variant of Figure 1). This source supplies a chosen voltage (for example, 6 to 10 V) so that M1 is turned on;

A Zener diode D3 is connected in opposition between the gate and source of M1, and a capacitor C2 to protect the MOSFET gate from excessively high or high-frequency voltages; and

a Zener diode D1 mounted opposite the gate of M1 and the drain, and with a resistor R3 and a diode D2 in the forward direction in the drain-to-grid direction limit the switching speed of M1; and

a circuit consisting of a (conventional) diode or a Schottky diode D4 mounted opposite each other on the drain of M1, and in series with a capacitor C1 and a resistor R1 connected to the positive terminal of the battery (4), so as to also limit the overvoltage when M1 is opened. In parallel, on the diode or the Schottky diode D4 there is mounted a resistor, for example fixed or variable, I1, connected, on the one hand, to the cathode of the diode and, on the other hand, to the drain of a second MOSFET M2 whose source is connected to the anode of the diode or the Schottky diode D4. The gate of M2 is controlled by an output of the analog or digital detection circuit, so as to prevent or cut off the charge or the charging current (Figure 7). The resistor I1 makes it possible to limit the charging current.

Figure 5 illustrates the first switching element (30) and the switch for discharge.

In this embodiment, the battery (4) is used by an external device (5), for example, and not limited to, by a starter motor characterized by the torque L1-R5.

Diode D4 is conductive when the battery (4) is discharging. It must withstand the short-circuit current for at least 10 ms and dissipate Joule losses during a high current discharge. In the case of a battery (4) consisting of an “8S1P” array of 8 individual cells in series, with an open circuit voltage of 26.4 V and an internal resistance of 0.1 ohm (including connections), the short-circuit current can reach 264 A.

When MOSFET M1 is turned on, an overvoltage greater than the drain-source voltage, Vds, of M1 may occur due to the cancellation of the current in inductor L1. D3 and C2 then act as a filter to protect the gate of MOSFET M1 from excessively high or high frequency voltages, and prevent it from being destroyed.

Zener diode D1, along with resistor R3 and diode D2, limits the switching speed of M1 so that it does not oscillate. Switching (in voltage) is defined by the opening and closing of MOSFET M1, i.e., its off state and its on state.

In certain embodiments, the disconnection device (3) comprises a second optocoupler switching device (31) in which

a first MOSFET M1 connected by its source to the negative terminal of a set of individual elements, the gate of this MOSFET M1 is controlled by a voltage, this source supplying a voltage chosen so that M1 is always on,

a circuit consisting of a (conventional) diode or a Schottky diode D4 mounted opposite each other on the drain of M1, in parallel on the Schottky diode D4, a fixed resistor, I1, is mounted, connected, on the one hand, to the cathode of the Zener diode D4 and, on the other hand, to the drain of a second MOSFET M2 whose source is connected to the anode of the Schottky diode D4, the gate of M2 being connected to the positive terminal of said set of individual elements,

a Zener diode D6 and a resistor R6 in series with the gate of M2, with the Zener diode D6 mounted in the forward direction in the drain-gate direction,

a Zener diode D5 mounted in opposition between the gate and the source of M2 to define, with the Zener diode D6, the value of the voltage at the gate of M2, and at which M2 is turned on,

a capacitor C5 connected between the gate and source of M2, and in parallel with Zener diode D5, to protect the gate of M2 from high frequency voltages,

an optocoupler OP1 mounted between the gate and source of M2, and in parallel with capacitor C5 to block M2 in case the voltage or temperature of an individual cell element or set of cell elements is exceeded, then MOSFET M2 cutting off the load current.

The appropriate value of the fixed resistor I1 can be obtained by replacing said resistor with a variable resistor and dynamically modifying its value (for example, and not limited to, by simulation).

Figure 6 illustrates the second switching device (30) and the load cut-off point.

In this embodiment, the battery (4) is charged by an external device (5), for example, and not limited to, an alternator or a charger characterized by the pair V4 - R5.

The external device provides, for example, but not limited to, 28 V in normal operation, but can output a higher voltage if its regulator fails. Standardized tests show a voltage 1.5 times the battery's nominal voltage, i.e., 42 V. In reality, this voltage can reach 80 V.

The charging current is limited by diode D4, which conducts in the discharge direction and is blocked in the charge direction, in parallel with a current-limiting resistor.

Diode D4 must therefore withstand the short-circuit current. This is feasible for a modular battery, but very difficult for a high-capacity battery. In fact, a non-modular 17 Ah battery, for example, can supply a short-circuit current of over 1800 A. Therefore, the diode must be able to conduct this current. Diodes that can conduct this current do not exist as "electronic components," and in practice, several lower-current components can be connected in parallel. However, balancing diode currents in parallel is almost impossible because the forward voltage drops sharply with temperature. Therefore, the hottest diode conducts all the current, which further increases its temperature, until it is destroyed.

MOSFET M1 is configured at its gate to be always on. Diode D4 is blocked during recharging, and the charge current flows through I1 and M2. I1 can also be a combination of resistors and polyswitches, or a semiconductor current regulator. M2 is on when its gate-to-source voltage V<gs> is 10 V, for example.

D6 is an 18 V Zener diode for example and D5 is a 10 V Zener diode. Therefore, under a voltage of 28 V (alternator or charger voltage), D6 and d 5 are at the limit of conduction, there is no current in R6 and V<gs>= 10 V. The voltage values of the diodes D5 and D6 therefore make it possible to define the value of the gate-source voltage of the MOSFET M2. C5 protects the gate of the MOSFET M2 from high frequency voltages.

The detection of a charger fault comprises two measurements: the measurement of the voltage of each cell and the measurement of the total voltage. If the voltage of one cell exceeds 4 V or if the total voltage exceeds 32 V, then the switching device (31) is actuated to switch to charging. Charging is possible again when the voltage drops below 26 V, for example due to a hysteresis comparator.

In the digital variant, the microprocessor will be connected to the battery to receive, at its inputs, both a voltage representative of the voltage of each cell element V<cec> constituting the battery, and also the measurement of the general voltage V<global> of the battery, available on the conductors leading to the external terminals of the battery. The program executable by the microprocessor will comprise a code module that monitors these two voltages V<cec> and V<global>; after comparing each one with a respective determined stored threshold, it activates the cut-off by activating the disconnection element (31) when this threshold is exceeded.

If the voltage or temperature of a single element is exceeded, optocoupler OP1 blocks M2. In fact, optocoupler OP1 comprises an LED (light-emitting diode) and a transistor. Therefore, if the voltage or temperature of a single element is exceeded, a current flows through the LED and causes the transistor to conduct. The gate-to-source voltage V<gs> of MOSFET M2 returns to near 0 V (V<cesat> for the collector-emitter saturation voltage of OP1). M2 then cuts off the load current (D4 and M2 are blocked).

The current in the OP1 LED is taken from the common point between M1 and the battery, therefore from the battery voltage (4), as shown in Figure 6, due to the disconnection between the 0 V of the battery and the 0 V of the alternator or charger.

Another alternative embodiment of the detection device is possible by using a low-pass RC circuit, nonlinear components (diodes and Zener diodes), and a comparator U1. An overvoltage detection assembly is provided around comparator U2.

It should be noted that this variant is also based on voltage comparison by using a divider bridge R9, R4 and a comparator U1 which sends the disconnection signal directly to the disconnection assembly shown in Figures 3 and 4. The assembly also uses the same supply principle by means of R7, D1, C2 of the comparators U1 and U2, to ensure detection even when the battery (4) voltage collapses due to a short circuit. The hysteresis assembly of U2 (Figure 2) for R5, R6, C3, is identical to that of Figure 1 for R6, R4, C4. Only the resistor R2 (Figure 2) is connected firstly to the common point of R5, C3 and to the positive terminal of the comparator U2, and secondly to the common point of R9 and R3.

The positive input of the hysteresis comparator U2 (Figure 2) is connected via a resistor R2 to the common point of R3, R9 and R7.

A Zener diode D2 is connected, on the one hand, to the positive input of the hysteresis comparator U2 and, on the other hand, via a resistor R1 to the common point of D1 and C2. The cathode of diode D2 is also connected via a capacitor C1 to the negative terminal of the battery (4) or to the modular set of elements.

In parallel with R9, a series assembly is mounted consisting of a resistor R3, a diode D3 with the cathode facing the positive terminal, and a Zener diode D4 with the cathode facing the common point of the divider bridge R9, R4. A capacitor C5 connects the common point of the bridge R9, R4 to the negative terminal of the battery (4) or of the cell or of the modular cell assembly.

In operation, as before, V1 represents the battery voltage (4). The reference voltage V2 at the positive input of comparator U1 is supplied by diode D2, which is a 16 V Zener. This diode is also connected via resistor R1 to the common point of D1 and C2. The cathode of diode D2 is also connected via capacitor C1 to the negative terminal of the battery (4) or to the modular array of elements. Diode D2 is therefore biased by R1 and filtered by C1. The battery voltage (4) is divided by R9 - R4, so that when the battery voltage (4) slowly reaches (in discharge) the end point voltage (or EPV, which is 18V for a 24V battery, and 9V for a 12V battery), the voltage at the negative input of U1 changes to 16V, which causes the comparator U1 to go from the “conducting” state to the “non-conducting” state (or from 1 to 0, in binary notation).

Below the end-point voltage, or EPV, the time constancy of R9-C5 causes U1 to switch with a typical delay of 1 to 30 s. The delay is shorter the further below the EPV the voltage is. For very low voltages, corresponding to a large overcurrent or a short circuit, R3, D3, and D4 are connected to R9 in parallel, reducing U1's switching time to less than a second (typically 0.1 s). D4 is a 4.7 V Zener, for example.

Therefore, reading the features described above, the BMS offers:

A single device that simultaneously provides protection against short circuits, overcurrents and deep discharges;

Detection of an overcurrent for the activation of the cut-off by means of a voltage measurement;

A disconnection current that automatically adapts to the characteristics of the battery cells; no shunt, no magnetic sensor, no heating element. The disconnection curve is similar to a thermomagnetic curve, but with much greater precision (Figure 9);

The disconnection curve follows the aging of the storage elements;

No further adjustments are necessary.

Other advantages also result from the management system as described in this application, including, but not limited to:

very low static consumption;

circuits always on, no “on” or “sleep” mode;

a very high level of operational safety (high MTBF);

use of standard, non-strategic and non-specific components for the BMS;

For analog implementations: There is no software, no sequential logic, no clock, which avoids software problems and therefore battery deterioration in the event of a software problem;

No electromagnetic emission, and improved electromagnetic immunity (due to the use of low-power components). The present application describes various technical features and advantages with reference to the figures and/or the various embodiments. Those skilled in the art will understand that the technical features of a given embodiment may, in fact, be combined with the features of another embodiment, unless otherwise specified, or unless the combination does not provide a solution to at least one of the technical problems mentioned in this application. Furthermore, the technical features described in a given embodiment may be isolated from the other technical features of that embodiment, unless otherwise specified.

It should be apparent to those skilled in the art that the present invention allows for embodiments in many other specific forms, without departing from the scope of the invention as claimed. Accordingly, the present embodiments should be considered illustrative, but may be modified within the scope defined by the protection sought, and the invention should not be limited to the details given above.

Exhibit:

This corresponds to a non-limiting example of a digital integrator program to implement the integrator response in Figure 4B:

floatlastIntegratedValue = 0;

constfloat_SLOPE = -0.25;

constfloat_ORDINATE_ORIGIN = 2.5;

constfloat_COEF_PROGRESSIVITY = 1;

constfloat_VALUE_REF_INTEGRATION = 10;

constfloat_RAPID_THRESHOLD = 0.01;

constfloat_WEIGHTING = 1;

loop(){

floatGeneralvoltage = IO_Voltage(1) * 7;// retrieving the battery voltage that has been divided and rescaled lastIntegratedValue = Integration(generalvoltage, lastIntegratedValue);

if (lastIntegratedValue >= 1 ) Disconnection ();

}

floatIntegration (floatGeneralvoltage, floatlastValue){

if(generalvoltage<= _VALUE_REF_INTEGRATION){

if ((LastGeneralVoltage - generalVoltage)>_RAPID_THRESHOLD || (LastGeneralVoltage - generalVoltage)< -_RAPID_THRESHOLD)

_WEIGHT = 5;

else

_WEIGHT = 1;

float x = (_SLOPE * generalvoltage _ORDINATE_ORIGIN)*_WEIGHT;

float y = integration(x, _COEF_PROGRESSIVENESS);

int value = lastValue y;

return value;

}

return 0;

}

Claims (22)

i. Method for detecting abnormal operating conditions of an individual battery cell (4) or a plurality of the set of individual cells, comprising the following steps: sampling, at the common point of at least two resistors (R1, R2, R3, R4) of a divider bridge with at least two resistors (R1, R2, R3, R4), at least one voltage proportional to the voltage (V1) at the terminals of the individual element or of the set of individual elements; compare the detected voltage with a reference threshold; characterized in that said comparison with said reference threshold activates the implementation of: an assessment of future developments by means of an integral, evaluated analogically or digitally; a comparison of this evaluation with a detection voltage threshold Td from which a disconnection of the individual element or the set of individual elements monitored with respect to at least the battery terminals (4) is carried out. 2. Method according to claim 1, characterized in that the step of evaluation by means of digital integral comprises: steps to calculate the slope of evolution (P) of the voltage curve using at least two measurements made; at least one step to compare the calculated slope with a stored “RapidThreshold” value, i.e. if the slope exceeds the “RapidThreshold” value, applying a weighting coefficient that increases the acceleration of the evolution of the integral, so that it crosses the activation voltage threshold (Td) more quickly, or if it is not exceeded, a weighting coefficient with no acceleration effect. 3. Battery management system (1) (BMS) for accumulators, said batteries being able to be constituted by an individual element or several individual elements that can be arranged in a modular assembly and connected in series, in parallel or in a plurality of modular assemblies in series associated in parallel to form a battery (4), characterized in that said system (1) comprises suitable means for carrying out the steps of a detection method according to one of the preceding claims. 4. Battery management system (1) (BMS) for accumulators according to claim 3, characterized in that it comprises at least: a voltage divider bridge with at least two resistors (R1, R2, or R4, R9), for sampling at least one voltage proportional to the voltage at the terminals of the individual cell or set of individual cells of the battery (4), a detection device (2) for detecting abnormal conditions, a disconnection device (3), said detection device (2) communicates with the disconnection device (3) to activate it in case of detection of abnormal conditions, said disconnection device being able to be connected to the battery (4), and comprising at least two MOSFETs. 5. Battery management system (1) (BMS) for accumulators according to claim 4, characterized in that it comprises a microprocessor equipped with at least one storage memory allowing to store at least one “Refintegration” threshold variable and a stored detection voltage value (Td), the memory also containing the program executed by the microprocessor, allowing to collect the points of the voltage curve, the comparisons and decisions, the implementation of the equations allowing the integration, the microprocessor receiving as input the voltage (V<global>) coming from the common point of the divider bridge between a resistor (R1) and a resistor (R2), and storing the measurements according to a given frequency to observe the voltage curve (V<global>), and compare the values of the voltage curve (V<global>) with the “Refintegration” value, then, when the crossing of the “Refintegration” threshold is detected, said threshold being defined by the value stored in the memory, which activates the integration calculations of the curve. (V<global>)<l>, and compare the values of the calculated integration curve (V<integ.>) with a stored detection voltage value (Td) to activate the disconnection device (3) that performs the cut-off. 6. Battery management system (1) (BMS) for accumulators according to claim 5, characterized in that the storage memory of the microprocessor also comprises the value of a variable “RapidThreshold” stored in order to determine, by comparing the variation (dV) of the voltage (V<global>) between two successive instants (t1) and (t2), with the “RapidThreshold”, whether the calculation of the integral of the voltage curve (V<global>) must take into account a weighting coefficient or not. 7. Battery management system (1) (BMS) for accumulators according to claim 6, characterized in that the calculation of the integral comprises taking into account the variables of “Slope and/or Ordinary” calculated by the microprocessor from the data of the recorded voltage curve (Vglobal). 8. Battery management system (1) (BMS) for accumulators according to one of claims 3 or 4, characterized in that it comprises at least approximately a comparator (U1), a divider bridge (R1, R2, or R9, R4) mounted between the terminals of the modular assembly of the battery (4) or of an individual element of the battery (4) whose common point with the resistors is connected to the input of the negative terminal of the comparator (U1) to supply a voltage whose value is proportional to the value of the voltage (V1) at the terminals of the battery (4), in the ratio defined by the values of the two resistors (R1, R2, or R9, R4), and the positive terminal of the comparator is connected to a diode or to a supply cell to define the reference voltage (V2). 9. Battery management system (1) (BMS) for accumulators according to claim 8, characterized in that it comprises an integrating circuit around the comparator (U1), the integrating circuit comprising: a resistor (R5) connected between the common point of the divider bridge (R1), (R2) and the negative input of the comparator (U1), and a set of resistor (R8) and capacitor (C1), mounted in series by means of a common terminal, connected by the other terminal of (C1) to the output of the comparator (U1), the other terminal of (R8) being connected to the common point of the two resistors (R5, R8) and to the negative input of (U1), the values (R5) and (C1) being established for the disconnection intervention time, before the deterioration of the battery (4) in case of overcurrent detection. 10. Battery management system (1) (BMS) for accumulators according to claim 9, characterized in that a diode (D2) is connected in parallel to the resistor (R5), with its cathode connected to the common point of the divider bridge to change the integration time constant of the integrating circuit in case of overcurrent or short circuit. 11. Battery management system (1) (BMS) for accumulators according to one of claims 8 to 10, characterized in that the comparator (U1) has, at its output terminal, a voltage whose value characterizes a “non-conductive” state if the voltage applied to the input of the negative terminal of the amplifier (U1) is greater than the value of the reference voltage (V2), and a “conductive” state if the value of the voltage applied to the input of the negative terminal of the amplifier (U1) is lower than the value of the reference voltage (V2). 12. Battery management system (1) (BMS) for accumulators according to claim 4, characterized in that the detection device (2) comprises a capacitor (C3) mounted in parallel with (R2) which, combined with (R1), forms a filter for filtering out high frequency disturbances. 13. Battery management system (1) (BMS) for accumulators according to claim 8, characterized in that mounted in parallel with (R9) are a series assembly consisting of a resistor (R3), a diode (D3), with the cathode facing the positive terminal, and a Zener diode (D4), with the cathode facing the common point of the divider bridge (R9, R4), a capacitor (C5) that connects the common point of the bridge (R9, R4) to the negative terminal of the battery (4) or of the cell or of the modular assembly of individual elements. 14. Battery management system (1) (BMS) for accumulators according to claims 8 to 13, characterized in that a comparator circuit (U2) with hysteresis, arranged downstream of the comparator circuit (U1), comprises a set of hysteresis around the amplifier (U2) that receives, at the input of its negative terminal, the value of the voltage of the output of the amplifier (U1). 15. Battery management system (1) (BMS) for accumulators according to claim 14, characterized in that the hysteresis comparator (U2) comprises resistors (R3, R4) mounted as a dividing bridge between the positive and negative terminals of the battery (4) (V1), and whose common point to (R3) and (R4) is connected to the positive input of the comparator (U2) and a resistor (R6) that connects the output of (U2) to its positive input, to define the threshold and the hysteresis of the hysteresis comparator circuit comprising the amplifier (U2). 16. Battery management system (1) (BMS) for accumulators according to one of claims 8 to 15, characterized in that the detection device (2) comprises a resistor (R7) connected to the positive terminal of the battery (4), in series with a diode (D1), in the forward direction in normal operation, and a capacitor (C2) connected, on the one hand, to the cathode of the diode and, on the other hand, to the negative terminal of the battery (4) to allow it to be charged during normal operation; the supply inputs of the two comparators (U1), (U2) are connected to the common point with (D1) and (C2), this common point being used to maintain the supply of the amplifiers (U1) and/or (U2) when the voltage of the battery (4) collapses following a short circuit, to allow the disconnection to be triggered. 17. Battery management system (1) (BMS) for accumulators according to one of claims 8 to 16, characterized in that the reference voltage (V2) at the positive input of the comparator (U1) is provided by a Zener diode (D2), said Zener diode being connected via a resistor (R1) to the common point of (D1) and (C2), the cathode of the Zener diode of (D2) also being connected via a capacitor (C1) to the negative terminal of the battery (4) or to the modular set of elements. 18. Battery management system (1) (BMS) for accumulators according to one of claims 14 to 17, characterized in that the hysteresis comparator (U2) comprises a capacitor (C4, C3) connected in parallel to a resistor (R4, R3) and which, combined with another resistor (R3, (R2)), forms a filter for filtering out high-frequency disturbances and establishing a minimum disconnection time. 19. Battery management system (1) (BMS) for accumulators according to one of claims 14 to 17 or 18, characterized in that the positive input of the hysteresis comparator (U2) is connected via a resistor (R2) to the common point to (R3), (R9) and (R7). 20. Battery management system (1) (BMS) for accumulators according to one of claims 8 to 19, characterized in that the detection device (2) comprises a flip-flop connected to the output of (U1) or (U2), to store each action of the detection device (2) after each detection of deep discharges, overcurrent discharges and short-circuit discharges. 21. Battery management system (1) (BMS) for accumulators according to claim 4, characterized in that the disconnection device (3) comprises a switching device (30) in which a first MOSFET (M1) is connected by its source to the negative terminal of a set of individual elements, said MOSFET (M1) receiving, at its gate, the voltage source (V1) that controls (M1), said source supplying a voltage (V1) chosen so that (M1) is on, a Zener diode (D3), connected in opposition between the gate and the source of (M1), and a capacitor (C2) protect the gate of the MOSFET from excessively high or high frequency voltages, and a Zener diode (D1) mounted in opposition between the gate of (M1) and the drain, and with a resistor (R3) and a diode (D2) in the forward direction in the drain-grid direction limit the switching speed of (M1), and a circuit consisting of a Schottky diode (D4) mounted opposite to the drain of (M1) and in series with a capacitor (C1) and a resistor (R1) connected to the positive terminal of the battery (4) to limit overvoltage when opening (M1), in parallel with the Schottky diode (D4) there is mounted a fixed resistor (I1) connected, on the one hand, to the cathode of the diode and, on the other hand, to the drain of a second MOSFET (M2) whose source is connected to the anode of the Schottky diode (D4), the gate of (M2) being controlled by an output of the detection circuit, to prevent charging. 22. Battery management system (1) (BMS) for accumulators according to claim 4, characterized in that the disconnection device (3) comprises a second switching device (31) in which a first MOSFET (M1) connected by its source to the negative terminal of a set of individual elements, the gate of this MOSFET (M1) is controlled by a voltage, this source supplying a voltage chosen so that (M1) is always on, a circuit consisting of a Schottky diode (D4) mounted opposite to the drain of (M1), in parallel to the Schottky diode (D4) a fixed resistor (I1) is mounted, connected, on the one hand, to the cathode of the Schottky diode (D4) and, on the other hand, to the drain of a second MOSFET (M2), whose source is connected to the anode of the Schottky diode (D4), the gate of (M2) being connected to the positive terminal of said set of individual elements, a Zener diode (D6) and a resistor (R6) in series with the gate of (M2), the Zener diode (D6) being mounted in the forward direction in the drain-gate direction, a Zener diode (D5) mounted in opposition between the gate and the source of (M2), to define, with the Zener diode (D6), the value of the voltage at the gate of (M2), and at which (M2) is on, a capacitor C5 connected between the gate and the source of (M2), and in parallel with the Zener diode (D5), to protect the gate of (M2) from high frequency voltages, an optocoupler OP1 mounted between the gate and source of (M2), and in parallel with capacitor C5, to block (M2) in case the voltage or temperature of one element of the set of individual elements is exceeded, then cutting off the MOSFET (M2) the load current.